JPH0236320A - Abnormality diagnosing device for equipment - Google Patents

Abstract

PURPOSE:To detect the abnormality of equipment with high sensitivity by utilizing the change of an acoustic spectrum which occurs when the abnormality is caused in the equipment which is an object and detecting the abnormality of the equipment. CONSTITUTION:After a signal obtained by an acoustic sensor 1 is amplified by an amplifier 2 and converted into a digital signal by an A/D converter 4, it is divided into respective frequency components by a frequency analyzer 5. Only 6-12kHz signal components, out of the respective frequency component signals, are inducted into a computation device 6. After they are integrated, they are transformed to logarithmic display and outputted. The outputs are inducted into a 1st and 2nd comparators 7 and 8. Next, the comparator 7 outputs the value at the time when an input level exceeds a reference value and the comparator 8 outputs the value at the time when an input level exceeds a reference value obtained by storing value inputted one second before the present time as the reference value. As for the outputs from the comparators 7 and 8, whether or not the level difference is equal to or exceeding a certain value previously set is decided by decision devices 9 and 10 and the decided result is given to a display system driving device 11.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention provides two
The present invention relates to an equipment abnormality diagnosis device that can remotely diagnose equipment abnormalities.

(Prior Art) When an abnormality occurs, many devices generate a sound different from that during normal operation. From such a thing.

By monitoring the sounds generated by two conventional devices.

BACKGROUND ART A device abnormality diagnosis device that can remotely diagnose a device abnormality is being considered. An abnormality diagnosis apparatus for such a device usually includes an acoustic sensor provided near the device and a signal processing device for processing the output signal of the acoustic sensor at a location remote from the acoustic sensor. The signal processing device is
When the overall level of the output signal of the acoustic sensor exceeds a certain level, it is determined that there is an abnormality and a message is output.

However, the conventional equipment abnormality diagnosis apparatus configured as described above has the following secondary problems. That is, the conventional device monitors the overall level of the output signal of the acoustic sensor. In this type of system, even if it seems obvious that an abnormality has occurred when heard, no abnormality message is issued if the level difference itself is small. Therefore, there was a problem of lack of reliability.

Additionally, there is a problem in that abnormal messages are likely to be issued in response to external noise such as a public address system, leading to frequent malfunctions.

(Problems to be Solved by the Invention) As described above, conventional device abnormality diagnosis devices have a problem in that they have low detection sensitivity to abnormal sounds and may malfunction due to external noises.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an equipment abnormality diagnosing device that can improve the detection sensitivity for abnormal sounds, and an additional object of the present invention is to provide an equipment abnormality diagnosing device that is less likely to malfunction due to external noise. It is said that

(Means for Solving the Problems) In order to achieve the above object, an apparatus for diagnosing equipment abnormality according to the present invention has an improved signal processing device. That is, the signal processing device includes a frequency analyzer that divides the output signal of the acoustic sensor into each frequency component, and a means for extracting only the signal component in a pre-designated frequency band from each signal component divided by the frequency analyzer. , means for determining an abnormality when the combined level of the extracted signal components exceeds a predetermined level.

Further, the signal processing device is configured to process only the output signals continuously sent out from the acoustic sensor during a period in which the level is above a certain level and the duration is within a certain period. There is.

(Function) When an abnormality occurs in the equipment, the sound emitted from the tumbler changes. The frequency band of this changing sound is often determined by the device. For example, in thermal power plants, etc.
When steam leaks from the steam guide pipe, most of the steam leak sound is dominated by high frequency components of 6 KHz or higher. Namely.

When a steam leak occurs, the sound level above BKHz increases rapidly. Therefore, if you set in advance to extract only the signal components of 6 kHz or higher from each signal component separated by the two-frequency analyzer, when a steam leak occurs, the combined level of the signal components in this band will increase rapidly. , This makes it possible to detect that a steam leak has occurred. in this case.

It does not respond to sounds below 8 kHz, that is, external noise. This makes it possible to detect steam leaks, that is, equipment abnormalities, with high sensitivity.

In addition, the signal processing device is configured to process only those output signals that are continuously sent out from the acoustic sensor and that have a level above a certain level and a duration that satisfies the conditions that the duration is within a certain level. When the frequencies of the abnormal sound and the disturbance sound are close to each other, it is possible to select the abnormal sound and respond only to the abnormal sound. For example, if the duration of the abnormal noise is relatively short, such as the sound of hammering from piping, it is possible to distinguish whether the abnormal noise is from equipment or an external disturbance by monitoring the duration.

(Example) An embodiment will be described below with reference to one drawing.

FIG. 1 is a block diagram of an apparatus for diagnosing an abnormality in equipment according to an embodiment. This embodiment is an example in which the present invention is applied to detect steam leakage from steam guide pipes constituting a plant.

In the figure, 1 indicates an acoustic sensor such as a microphone, and 2 indicates an amplifier that amplifies the output signal of the acoustic sensor. The acoustic sensor 1 and the amplifier 2 are installed in a ring above the floor of the building in which the steam guide pipe is installed. Reference numeral 3 indicates a signal processing device, and this signal processing device 3 is provided at a location away from the location where the acoustic sensor 1 is installed, for example, in a central control room.

The signal processing device 3 is configured as follows.

That is, after the output signal of the amplifier 2 is converted into a digital signal by the A/D converter 4, this signal is converted to a digital signal by the high-speed frequency analyzer 5.
The frequency component signal is divided into each frequency component signal up to 12 kHz. Then, each frequency component signal divided into two is introduced into the arithmetic unit 6. As shown in FIG. 2, this calculation device 6 first calculates a power spectrum (step SL) L,
This is repeated multiple times to average the data (step S2). Next, integrate the data from 6 to 12 K11z (
Step S3) After

This value is converted into logarithmic representation (step S4) and output. This output is introduced into a first comparator 7 and a second comparator 8, as shown in FIG. A reference value for comparison is stored in the comparator 7 in advance. 1. When the input level exceeds the reference value, this comparator 7 outputs the exceeded value. on the other hand.

The comparator 8 stores an input value one second before the current time as a reference value for comparison. When the two-person power level exceeds the reference value, the comparator 8 outputs the exceeded value. The outputs of these comparators 7 and 8 are determined by determiners 9 and 10, respectively.
given to. The determiner 9.10 determines whether the level difference obtained by the comparator 7.8 is a predetermined constant value (for example, 3 dB).
) or more, the output is sent. These outputs are given to one display system driving device 11. The display system drive device 11 is configured to, when receiving any of the output signals of the 1 judgment 19 and 10, display the frequency analysis diagrams before and after receiving this signal on the display 12, and also to sound the alarm 13. There is.

With this configuration, if a steam leak occurs from the steam guide pipe, this steam leak can be reliably detected without being affected by external noise. In other words, when a steam leak occurs, the acoustic sensor As an example, the acoustic spectrum within the premises obtained from the output signal of No. 1 is shown by the broken line in Fig. 3.On the other hand, the acoustic spectrum in normal conditions is shown by the solid line in Fig. 3. As can be seen, at the overall level, the components below 2KHz are dominant, and their levels hardly change even if a steam leak occurs.However, when looking at the components above BKHz, the levels change significantly.

In this embodiment, the signal obtained by the acoustic sensor 1 is divided into frequency component signals by the frequency analyzer 5, and only the signal components of 6 to 12 KHz among the divided signal components are detected, that is, whether steam leakage occurs or not. The comparator 7 compares the synthesized level with the reference level by extracting only the signal components in the frequency band whose level changes significantly when Drive device 11
I'm trying to get it to work. Therefore, unlike conventional devices that monitor at an overall level, steam leaks can be detected with high sensitivity. Also,
The comparator 8 compares the level with the level one second ago, and when the difference is greater than a certain value, the determiner 10 outputs an output.

It is also possible to quickly detect whether the human power level exceeds the reference value stored in the comparator 7 or whether an abnormality is gradually progressing.

- As can be seen from the operation described in F, if a plurality of signal processing systems below the two-frequency analyzer 5 are provided corresponding to each device, abnormalities in each device can be detected with high sensitivity.

FIG. 4 is a block diagram of a device abnormality diagnosis apparatus according to another embodiment. This embodiment is an example applied to detecting the hammering sound of piping. In this figure, the same parts as in FIG. 1 are indicated by the same reference numerals.

In the device abnormality diagnosis device according to this embodiment.

The magnitude of the signal at the next stage of the A/D converter 4 is constant and is less than or equal to Bell.
A level determiner 21 is provided that allows the signal to pass only when -, and a duration determiner 22 is provided that allows only the signal that has passed the level determiner 21 and whose duration is within a certain time (for example, 0.6 seconds) to pass. This duration determination device 22 is provided.
The signal that has passed through is introduced into a high-speed frequency analyzer 5. Further, after calculating the power spectrum (step S1) as shown in FIG.
．． Integrate signal components up to 0KHz (step S2) L
, This is converted into logarithmic representation (step S3) and output. Further, the comparator 7 stores in advance a reference value for comparison. Furthermore, the comparator 8.

The human power level 0.2 seconds before the current time is stored as a reference value for comparison.

With such a configuration, the hammering sound of the piping can be detected without being affected by a public address system or the like. In other words, the hammering sound of piping is usually 0.5 to 1.
The 2 KHz signal component occupies most of the signal components, and the duration is 0.4 to 0.6 seconds or less, as shown in FIG. On the other hand, human voices such as public address systems, etc. are 9
The frequency band is 0.5 to 2.0 KHz, which is close to the frequency band of hammering sound.

However, as shown in FIG. 7, the duration of a horizontal broadcast is rarely less than 0.6 seconds.

In this embodiment, the level determiner 21 extracts only signals above a certain level from the signals output from the A/D converter 4,
Further, from among the extracted signals, a duration determiner 22 extracts only signals whose duration is 0.6 seconds or less, and this signal is divided into each frequency component signal by a frequency analyzer 5.
0.5 to 2.0 K of each separated signal component
Only the Hz signal component.

In other words, only the signal components in the frequency band of the hammering sound are extracted, and the synthesized level and the reference level are compared by the comparator 7, and when the difference is greater than a certain value, the determiner 9 determines that there is an abnormality, and the display system drive device 11 I'm trying to get it to work. Therefore, even in a frequency band close to the hammering sound, there is no response to a sound with a long duration like a public address system, that is, a disturbance sound, and the hammering sound can be detected with high sensitivity. Also, in the comparator 8, it is compared with the level 0.2 seconds before, and when the difference is more than a certain level, it is determined that n
Since the output is sent from the comparator 7, the input level does not exceed the reference value stored in the comparator 7.
It is also possible to detect a state in which an abnormality is gradually progressing.

[Effects of the Invention] The equipment abnormality diagnosis device according to the present invention configured as described above detects an abnormality in the target equipment by using a change in the acoustic spectrum that occurs when an abnormality occurs in the equipment. This makes it possible to detect abnormalities in the equipment with high sensitivity.

Furthermore, by monitoring the duration, the influence of external noise can be removed and malfunctions can be reduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a device abnormality diagnosis device according to an embodiment, and FIG. 2 is a diagram showing the flow of data processing in an arithmetic unit incorporated in the device. Fig. 3 is a diagram showing a comparison of the acoustic spectrum in the premises when steam is leaking from the steam guide pipe and the acoustic spectrum under normal conditions, and Fig. 4 is a block diagram of an equipment abnormality diagnosis device according to another embodiment. Configuration diagram. Fig. 5 is a diagram showing the flow of data processing in the arithmetic unit incorporated in the same device, Fig. 6 is a diagram showing a comparison of acoustic spectra in the premises with and without a public address system;
FIG. 7 is a diagram showing an example of the duration of a public announcement, and FIG. 8 is a diagram showing an example of the duration of the piping hammering sound. 1... Acoustic sensor, 3, 3a... Signal processing device, 5
...Frequency analyzer, 6.6a...Arithmetic device, 7,8
... Comparator, 9.10 ... Judgment device, 11 ... Display system drive device, 12 ... Display device, 13 ... Alarm device, 2
1...Level determiner, 22...Duration time determiner. Applicant's representative Patent attorney Takehiko Suzue Figure 1 Arrival Figure IE, Ordinary B sound frequency mantissa (in Hz) Figure ■ Question (sec) Figure 8 Morima (sec) Figure

Claims (2)

[Claims] (1) Abnormality in a device that is equipped with an acoustic sensor placed near the device and a signal processing device that is placed away from the acoustic sensor and determines whether the device is abnormal based on the output signal of the acoustic sensor. In the diagnostic device, the signal processing device includes a frequency analyzer that divides the output signal of the acoustic sensor into each frequency component, and a frequency analyzer that divides the output signal of the acoustic sensor into each frequency component, and only the signal component in a prespecified frequency band among the signal components divided by the frequency analyzer. An apparatus for diagnosing an abnormality in equipment, comprising: means for extracting signal components; and means for determining an abnormality when a combined level of the extracted signal components exceeds a predetermined level. (2) The signal processing device processes only those output signals that are continuously sent out from the acoustic sensor and have a period that satisfies conditions in which the level is above a certain level and the duration is within a certain level. An abnormality diagnosis device for the device according to item 1.